Using Structure in Content-Based Image Retrieval.
Iqbal Qasim, J.K. Aggarwal · 1999
In this paper we present a study of the comparison of the performance of content-based image retrieval systems based on structure [1] with those based on histogram and texture analysis methods, where retrieval is concerned with locating images containing manmade objects. The advantage of using structure in such queries is demonstrated by analyzing an image database containing monocular grayscale outdoor images taken from a ground-level camera to retrieve images containing buildings. Keywords – Content-based image retrieval, histogram, texture, structure, nearest neighbor classifier, Bayesian formulation. 1